Re: Efficiency of light sources.

On Jun 5, 7:37 pm, Rich <some...@xxxxxxxxxxxxx> wrote:
[...]

Opps, the biggest problem today would probably be the cost. You
can buy such bulbs today for about $65.

Solid state lasers can be rather short lived and very
fussy about over currents.

While true I was not referring to solid state lasers, they tend
to be monochromatic, pinpoint and make very poor light sources.

Understood. My point was that the high current, often pulsed,
required to get them to ~laze~ may be comparable to that in
the high power LEDs, making them much less reliable than a
30ma panel indicator. I don't know that to be a fact. But you
can probably find out *before* you take out a second mortgage
on the family pet to relamp the house.


And one low-power devices used in DVD players seem to last quite
a long time. Now if only the write lasers in DVD burners lasted
longer.

You might research that
a bit before investing in a large quantity of the LEDs.

Well, I've already invested almost $10. This will ruin me! ;^}

Mortage your neighbor's pet. I won't tell anyone.


Typically I don't consider lighting costs as an investment, but
that's just me.

Here is a typical driver for power LEDs with current limiting:
http://www.maxim-ic.com/appnotes.cfm/appnote_number/3532

I don't think low-cost lighting typically uses ICs, although I
could be mistaken. Looking into my LED nightlight I see what looks
like 4 diodes and a resistor. No doubt there's a small transformer
underneath. Looks like a transformer, a full-wave bridge rectifier
and a pulldown resistor. Simple, seems to work.

So the more current limiting they design in,
the lower the effieciency.

Maxim might tell ya who buys their regulators and why.
That is a cheap survey that won't even require pawing
your wedding ring. :o)


A recent Scientific American had an article on plasmonics, which
turned out to be very interesting. It offers two methods of
increasing the light output of LEDs, one method I recall stated
about 10X the light output and they expect from both methods combined
about 100X the light output. So, in theory at least, the intensity
problem has either been solved, or is solvable.

Anybody have any comments? if they had 100W LED bulbs today for
$20, would anybody buy them? How about $5?

The would also, I assume, work in the cold where fluorescents are
slow to start or won't start a'tall. And they start instantly.

Yes... I have a few screw base type flourescents that are
getting cold natured after about 4 years.

Geez, what kind are you using? I finally started recording the
start and replace dates of compact fluorescents and with 5 replaced
already this is what I find.

Min 1.08 (years)
Max 2.08 (years)
Avg 1.60 (years)

The average went down from 1.7 years after I replaced one of
the bathroom CF's this morning. Note, I was lazy and used a
365 day year. I don't think I've ever seen a CF bulb last 4
years.

No mfr on the one I just retired but I am careful to put them where
they can run continuous or nearly continuous. Besides maximising
the investment I was drawing from some studies that show building
managers where the breakpoint is to balance power savings with
shorter lamp life due to cycling. My forgetter is in turbo boost
but I was really shocked how low the threshold was. IOW if on for
6 hours you may as well leave it on all day. Particularly in an office
building where the lamps are considered part of the heaing system.

This table seems consistant with the notion:
<< Lifetime: the lifetime of fluorescent tubes and lamps
depends greatly on their mean "on-time" (i.e. how long they
are emitting light for on average). >>
http://www.epanorama.net/documents/lights/neonlamps.html

Sue...


Cheers,

Rich



Sue...


.

Relevant Pages

  • Re: Digital Current Control for LED array
    ... >>> voltage across them assuming they are of the same type. ... >>light output levels due to differences in phosphor coating (for white ... >>LEDs) thermal conductivity of the package and other physical ... A parallel connection sets the Voltage of each LED as fixed and NOT the current. ...
    (sci.electronics.design)
  • Re: Stocked up with 100w bulbs yet?!
    ... it seems that if you power them with a ... persistence of vision trick to interpret light when there is none. ... other technical reason to drive the LEDs such that they flash. ... average current as indeed is the light output. ...
    (uk.legal)
  • Re: White LED light output vs current
    ... discovered that the light output increases nicely up to ... find LEDs that don't increase in brightness up to at least a few tens ... LEDs drift to long wavelengths with temperature. ... boat (British 'narrow boat') with a 12V DC system. ...
    (sci.electronics.design)
  • Re: CFL Color Temperature
    ... There's very little/none light output decay with either straight LEDs or ... LEDs with phosphor; ... LED life is now most ... light output depreciation over life. ...
    (sci.engr.lighting)
  • Re: Correct way to drive a pulsed LED
    ... >> I'm planning on desgining a refresh controller that handles one or two ... pixels wide for each character, 22 characters per line, 9 lines ... 880 LEDs if all are lit. ... and 1/11 the light output is "the same". ...
    (sci.electronics.basics)
Quantcast